Physical measures and biomarkers in the Health and Retirement Study

The collection of physical measures and biomarkers has become increasingly common in population-based survey research. The logistics of collecting these measurements in a national survey are complex and approaches have differed across surveys and organizations. The National Health and Nutrition Examination Survey (NHANES) provides one model, in which detailed clinical measurements are conducted by medical professionals in an examination center. The Health and Retirement Study (HRS) is another model, in which measurements are collected in respondents' homes by lay interviewers. Although the HRS physical and biological measurements are less extensive than those in NHANES, the HRS has some advantages in that it has a much larger sample size than NHANES for the 55+ age group, is longitudinal, and collects much more detail on SES, health care utilization, and family support.

HRS began experimenting with collecting physical measures and biomarkers in pilot projects beginning in 2003, culminating in a full-scale implementation in the 2006 wave. The measurements conducted include grip strength, lung function, balance stands, walking speed, anthropometric measurements (height, weight, waist circumference), as well as blood pressure, dried blood spots, and salivary DNA. Completion rates ranged from 93% for the physical measurements to 81% for the blood spots, with some variation by age, race/ethnicity, education, and health status. This paper provides an overview of the motivation for collecting physical measures and biomarkers in the HRS, the implementation of the new field protocols, completion rates by key demographic and health covariates, and some preliminary substantive results and comparisons with other surveys.